D.J. ALLEN l , J.M. LENNE 2 and D.WOOD
l Higher Quantock, Stockland, Honiton, Devon, EX14 9DX, UK; 2 13 Herons Quay, Milnthorpe, Cumbria LA7 7HW, UK.

Background and objectives
Gene centres of plants have long been recognised as rich sources of resistance against diseases where host and pathogen have co-evolved, perhaps in a gene-for-gene relationship. Evidence for such co-evolution in common centres of origin in a long established relationship is well illustrated by recent work on the common bean ( Phaseolus vulgaris) and its angular leaf spot pathogen ( Phaeoisariopsis griseola . Among the world's major crops, legumes are notorious for the degree of seed transmissibility of their pathogens, so that many of their co-evolved pathogens are distributed worldwide with their hosts through crop introduction, which is a major feature of modern agriculture. Nevertheless, it is sometimes argued that greater productivity of many crops occurs outside centres of origin, suggesting a lesser diversity of parasites in the 'new' continent. This ignores one crucial factor : as crops often originated in one continent as members of large genera with much wider distribution, introduction to other regions could place the crop in contact with geographically distant but taxonomically related wild species and their pathogens, with the emergence of new encounter diseases [1] which, we argue, are a great deal more common and important than is generally recognised. Our objective is to draw fresh attention to the concept of new encounter in the origin of epidemics with reference to crop introduction and the search for host plant resistance.

Results and conclusions
A search of the literature reveals that new encounters are especially common among virus diseases, with examples including cocoa swollen shoot, African cassava mosaic, maize streak, rice yellow mottle, groundnut rosette and common bean black root, all in Africa in crops of American origin. Rice hoja blanca and Indian peanut clump are examples of new encounters in America of an Asian host and in Asia in an American host, respectively. Moko disease of banana ( Burkholderia solanacearum ) and, on present evidence, perhaps also halo blight of common bean (caused by races 4 and 8 of Pseudomonas syringae pv. phaseolicola ) are examples of new encounters with bacterial pathogens. Among fungal diseases, red leaf blotch of soyabean ( Dactuliochaeta glycines ), cowpea stem rot ( Phytophthora vigna ) and bean scab ( Eisinoe phaseoli ) are examples of new encounters in Africa of an Asian plant, in Australia of an African plant, and in Africa of an American plant, respectively. It is sometimes found that secondary centres of genetic diversity of a crop are useful hunting grounds for resistance against new encounter diseases. The extent to which resistance has evolved in such areas presumably relates to several factors including the time of first introduction of the crop, the time of the new encounter, the breeding system of the crop, the presence of wild relatives with which the crop might naturally introgress, and the environment. Surprisingly though, there is an increasing number of cases wherein resistance is found in germplasm from areas free of the disease, a type of resistance best considered allopatric [2], in which there is an association between resistance to the new encounter disease and some other character. An important source of resistance in maize to streak is cv. Revolution which comes from Reunion where streak is thought not to occur. Similarly, new sources of resistance to rosette have been found among groundnut landraces in India where they have never been exposed to the disease. In common bean scab, resistance has been identified in germplasm freshly introduced from Latin America where the disease does not occur. We discuss possible reasons for the occurrence of allopatric resistance.

1. Buddenhagen, IW, 1977. Annals of the New York Academy of Sciences 287, 309-326. 2. Harris KM, 1975. Environmental Entomology 4, 661-669.